28 Mr. Woods on the Genera of European Grasses. 



outer palea of the perfect floret being always within the outer glumes. The 

 glumes are large, nearly equal, completely inclosing the florets. I have ad- 

 mitted amongst them Zea and Lygeum without being able to give a good 

 reason for doing so. The latter especially seems to me to have little relation- 

 ship to the rest of the tribe ; but I follow Kunth, since I know not where to 

 place them better, 



1. Phalaris. Glumes navicular, carinate, herbaceo-membranous, including. 



External florets, scale-like, neuter, Palese coriaceous, closely investing 

 the seed, 



2. Anthoxanthum. Glumes membranous, including. External florets neuter, 



each of one large palea, both of them awned. Perfect floret diandrous. 

 Palese scariose. 



3. Hierochloa. Glumes membranous, thin. External florets triandroiis. Per- 



fect floret diandrous, Paleae firmer than the glumes, 



4. Zea. Barren florets 2 in a spicula, disposed in a terminal panicle. Fertile 



floret 1 in each spicula, with external abortive rudiments disposed in a 

 dense lateral spike, 



5. Lygeum. Glumes 0, Spiculae all fertile, in pairs or threes, uniting at last 



to form a 2- or 3-seeded nut. 



Phalaris has in most species two small fleshy scales at the base of the per- 

 fect floret, often tipt with hairs, and I think I can sometimes discern traces 

 of a filmy interior palea. In P. canariensis the additional floret is larger and 

 more like the paleae of the perfect floret both in size and substance than it is 

 in any other European species, 



I have no hesitation in considering the parts which used to be described as 

 the corolla of Anthoxanthum as abortive florets, because I think the presence 

 of a midrib and central awn a very strong proof that the upper of these cannot 

 be an interior palea. This genus and Hierochloa I am quite disposed to 

 place with the Phalaridece ; but I find it more difficult to say why Arrhena- 

 THERUM should not also be placed amongst them. It has, however, only one 

 barren floret, without any rudiment or indication of a second, and the fertile 

 floret is consequently placed within the superior glume as in the Panicecp. 

 Kunth attributes to it the filiform rudiment of a superior floret, which is not 



